Amir Hossein Abdollahi Nohoji, Parviz Keshavarzi, Mohammad Danaie
{"title":"采用Ge2Sb2Te5相变材料制备的低功耗全光光子晶体突触","authors":"Amir Hossein Abdollahi Nohoji, Parviz Keshavarzi, Mohammad Danaie","doi":"10.1016/j.rio.2025.100901","DOIUrl":null,"url":null,"abstract":"<div><div>In this study, we propose an all-optical neuromorphic photonic crystal synapse structure equipped with a Ge<sub>2</sub>Sb<sub>2</sub>Te<sub>5</sub> phase-change material (GST-PCM). By leveraging the unique properties of the GST-PCM material, this structure enables the control of the synaptic weight through targeted and focused laser irradiation. 3D simulations employing the finite-difference time-domain (FDTD) and finite element method (FEM) demonstrated optical transmission exceeding 99 % and reflection below −20 dB at a wavelength of 1504 nm within the proposed structure. These features, together with their compact dimensions and low power consumption, make our proposed structure an ideal candidate for optical processing applications and neuromorphic neural networks. Furthermore, we investigated the physical and thermal equations to determine the crystallization fraction of GST-PCM during the synapse weighting process. These equations showed excellent agreement with the simulation results and could accurately calculate the GST-PCM crystallization fraction as a function of time and laser power. Our proposed structure not only has the potential to be extended to neuromorphic systems and optical neural networks but also serves as an innovative platform for all-optical synapses because of its precise control of optical properties, high adaptability, and low power consumption.</div></div>","PeriodicalId":21151,"journal":{"name":"Results in Optics","volume":"21 ","pages":"Article 100901"},"PeriodicalIF":3.0000,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Low-power all-optical photonic crystal synapse using Ge2Sb2Te5 phase-change material\",\"authors\":\"Amir Hossein Abdollahi Nohoji, Parviz Keshavarzi, Mohammad Danaie\",\"doi\":\"10.1016/j.rio.2025.100901\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>In this study, we propose an all-optical neuromorphic photonic crystal synapse structure equipped with a Ge<sub>2</sub>Sb<sub>2</sub>Te<sub>5</sub> phase-change material (GST-PCM). By leveraging the unique properties of the GST-PCM material, this structure enables the control of the synaptic weight through targeted and focused laser irradiation. 3D simulations employing the finite-difference time-domain (FDTD) and finite element method (FEM) demonstrated optical transmission exceeding 99 % and reflection below −20 dB at a wavelength of 1504 nm within the proposed structure. These features, together with their compact dimensions and low power consumption, make our proposed structure an ideal candidate for optical processing applications and neuromorphic neural networks. Furthermore, we investigated the physical and thermal equations to determine the crystallization fraction of GST-PCM during the synapse weighting process. These equations showed excellent agreement with the simulation results and could accurately calculate the GST-PCM crystallization fraction as a function of time and laser power. Our proposed structure not only has the potential to be extended to neuromorphic systems and optical neural networks but also serves as an innovative platform for all-optical synapses because of its precise control of optical properties, high adaptability, and low power consumption.</div></div>\",\"PeriodicalId\":21151,\"journal\":{\"name\":\"Results in Optics\",\"volume\":\"21 \",\"pages\":\"Article 100901\"},\"PeriodicalIF\":3.0000,\"publicationDate\":\"2025-09-25\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Results in Optics\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S2666950125001294\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"Physics and Astronomy\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Results in Optics","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2666950125001294","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"Physics and Astronomy","Score":null,"Total":0}
Low-power all-optical photonic crystal synapse using Ge2Sb2Te5 phase-change material
In this study, we propose an all-optical neuromorphic photonic crystal synapse structure equipped with a Ge2Sb2Te5 phase-change material (GST-PCM). By leveraging the unique properties of the GST-PCM material, this structure enables the control of the synaptic weight through targeted and focused laser irradiation. 3D simulations employing the finite-difference time-domain (FDTD) and finite element method (FEM) demonstrated optical transmission exceeding 99 % and reflection below −20 dB at a wavelength of 1504 nm within the proposed structure. These features, together with their compact dimensions and low power consumption, make our proposed structure an ideal candidate for optical processing applications and neuromorphic neural networks. Furthermore, we investigated the physical and thermal equations to determine the crystallization fraction of GST-PCM during the synapse weighting process. These equations showed excellent agreement with the simulation results and could accurately calculate the GST-PCM crystallization fraction as a function of time and laser power. Our proposed structure not only has the potential to be extended to neuromorphic systems and optical neural networks but also serves as an innovative platform for all-optical synapses because of its precise control of optical properties, high adaptability, and low power consumption.